On 11/14/2013 12:58 AM, Charles Martin wrote:
I was hoping to find a detailed discussion of the binary model format and
the scoring function
There are two papers describing the maxent and perceptron:
A Simple Introduction to Maximum Entropy Models for Natural Language
Processing, by Adwait Ratnaparkhi
http://repository.upenn.edu/cgi/viewcontent.cgi?article=1083&context=ircs_reports
Discriminative training methods for hidden markov models: Theory and
experiments with perceptron algorithms, by Michael Collins
http://dl.acm.org/citation.cfm?id=1118694
The only good documentation for the binary format are comments in the
source code.
HTH,
Jörn
